Which test is the most effective for tuning PID controllers?

PID loop tuning isn't always straightforward. In fact, appropriate tuning of a PID controller entails a number of processes that will result in consistent outcomes and optimal controller performance. Testing is, without a doubt, the most crucial phase. A test can reveal the dynamic relationship that exists between a controller's process variable (PV) and its controller output when done correctly (CO). The results of such a test enable for the creation of a precise model and the selection of tuning parameters that will best manage the operation.

Determining which of the many testing choices is best for their case is a common dilemma for practitioners. The answer is, as is so often the case in the field of automation and process control, it depends. To begin, how quickly does the loop react to changes? How sensitive is the content being processed to change? How much time do you have when testing to make changes? These and other factors are crucial in determining which test is right for you.

The four tests listed below are used to reveal dynamic behaviour and fine-tune PID control loops. The following fundamental advantages and disadvantages might help you decide which test is ideal for your PID controller tuning needs:

Step-by-step evaluation

The step test may appear to be the easiest of all the methods, but it has a few noteworthy drawbacks. The CO is "stepped" from one constant number to the next in this test. It should cause the PV to move from one steady state to another steady state, and the size of the change should be substantial enough to identify the response from any noise in the process. If modelling and tuning software isn't accessible, the step test may be the only viable choice, as model parameters can be manually derived using the test data's trend. The step test, on the other hand, is often time-consuming because a steady state is necessary at both the start and finish of the test. Furthermore, any data obtained during testing may be invalidated if there are any disruptions.

Pulse check

The pulse can be thought of as two step tests that are conducted in sequence, in opposing directions, and without the need for a steady-state in between. It begins by adjusting the CO value and moving the PV in one direction until a definite response is observed. Once the PV response has been established, the CO is moved in the opposite direction — often before the process has reached a new steady state. CO is eventually restored to its original steady-state value. This method is more detailed than the step test since it reverses the process, exposing various dynamic behaviour within the process (e.g. heating vs. cooling). The pulse, on the other hand, necessitates the use of software to precisely determine model parameters.

Test with a doublet

The doublet test is the most comprehensive and least disruptive of the four alternatives presented here. On the surface, there appears to be two pulses, one after the other and in opposite directions. It's worth noting that the doublet test has a limited impact on the process because the first and third PV changes are tiny and often smaller than those required for a single pulse test. The difference-maker in the doublet test is the second PV change. The second adjustment is twice as large as the previous two and spans the PV's initial steady-state value. The changes, taken together, provide a complete picture of the process's dynamic behaviour, exposing activity across a wider range of process areas.

PRBS

The pseudo-random binary sequence (PRBS) test is best suited for delicate operations that cannot withstand considerable or prolonged fluctuations in the PV value. A sequence of CO pulses with constant amplitude, alternate direction, and random duration is used. Each CO pulse pushes the PV slightly beyond the process's noise level. Individual pulses are smaller in size and create data that is only moderately functional in terms of tuning, but a group of PRBS pulses can produce highly functional data. This testing process is difficult in terms of negatives and requires a large amount of time. The PRBS, on the other hand, is a useful way of obtaining the data needed to improve PID loop performance for systems that cannot endure major changes.

Each of these testing methods is employed in industry, and each has distinct advantages and disadvantages. If the control loop can be kept at a constant condition at the start and end of testing, a manual step test has been found to give consistent results. The usage of software is advised for the other PID controller tuning choices mentioned in this essay. When it comes to tuning, most software recommends following a regular, repeatable approach, which is a good idea. Furthermore, despite high noise and oscillations, many software programmes have proven to calculate correct models.